Loss of ceramide kinase in Arabidopsis impairs defenses and promotes ceramide accumulation and mitochondrial H2O2 bursts

Abstract

Arabidopsis thaliana plants that lack ceramide kinase, encoded by ACCELERATED CELL DEATH5 (ACD5), display spontaneous programmed cell death late in development and accumulate substrates of ACD5. Here, we compared ceramide accumulation kinetics, defense responses, ultrastructural features, and sites of reactive oxygen species (ROS) production in wild-type and acd5 plants during development and/or Botrytis cinerea infection. Quantitative sphingolipid profiling indicated that ceramide accumulation in acd5 paralleled the appearance of spontaneous cell death, and it was accompanied by autophagy and mitochondrial ROS accumulation. Plants lacking ACD5 differed significantly from the wild type in their responses to B. cinerea, showing earlier and higher increases in ceramides, greater disease, smaller cell wall appositions (papillae), reduced callose deposition and apoplastic ROS, and increased mitochondrial ROS. Together, these data show that ceramide kinase greatly affects sphingolipid metabolism and the site of ROS accumulation during development and infection, which likely explains the developmental and infection-related cell death phenotypes. The acd5 plants also showed an early defect in restricting B. cinerea germination and growth, which occurred prior to the onset of cell death. This early defect in B. cinerea restriction in acd5 points to a role for ceramide phosphate and/or the balance of ceramides in mediating early antifungal responses that are independent of cell death.

Figure 1.

Subcellular Distribution of Arabidopsis Ceramide Kinase. (A) Subcellular localization of ACD5/ceramide kinase. The coding region of ACD5 was fused with the YFP at the C terminus, and then ACD5:YFP was coexpressed with a Golgi mCherry marker (ABRC stock number, CD3-968, row 1 panels) or an ER mCherry marker (TAIR, CD3-960, row 2 panels) by transient expression in Arabidopsis protoplasts. Protoplasts were examined by confocal microscopy 16 to 22 h after incubation. This experiment was repeated at least three times with similar results, and more than 20 protoplasts were analyzed at each time point (one protoplast is shown in each image). Bars = 10 μm. (B) and (C) Immunolocalization of ACD5 in wild-type plants. Arrowheads indicate the particles of immunogold-labeled ACD5. G, Golgi, M, mitochondrion; N, nucleus. Bar = 500 nm. (D) Immunoblot analysis of ACD5 in ACD5-RNAi (lines 7 and 8) and wild-type plants. Equal quantities of proteins extracted from 14-d-old plants were detected with an ACD5 antibody (upper panel). Coomassie blue staining of Rubisco (RBC) large subunit bands was for a loading control (middle panel). Protein levels were quantified as described in Methods (lower graph). Values are mean ACD5 levels relative to the wild-type control (set at 1). Error bars represent ±sd for three biological replicates. (E) Immunolocalization of ACD5 in ACD5-RNAi plants (line 8). Note that few particles of immunogold-labeled ACD5 were observed. Ch, chloroplast; M, mitochondrion. Bar = 500 nm. (F) Statistical analysis of ACD5 localization by immunoelectron microscopy in wild-type and ACD5-RNAi plants. Twenty mesophyll cells in each experiment were observed, and at least 30 photos were used for counting. Values are means ± sd. At least three independent experiments were done with similar results. Asterisks indicate significant differences between wild-type and ACD5-RNAi plants at P < 0.05 using Student’s t test. The percentage of gold particles means percentage of the particle numbers of each organelle in the total amount of particles.

Figure 2.

ROS Localization in Late Development of acd5 Plants and in Ceramide-Treated Protoplasts. (A) to (D) Forty-day-old wild-type leaves and acd5 mutant leaves with spontaneous cell death were pretreated with cerium chloride to visualize H₂O₂ production as described in Methods and were observed by electron microscopy without staining. At least three different leaf samples of both wild-type and acd5 plants were cut in each experiment. Images represent typical observations in two independent experiments. Bars = 500 nm. (A) Wild-type mesophyll cells. Note that all organelles are cerium-free. (B) to (D) acd5 mutant mesophyll cells. Note electron-dense cerium deposits (arrowheads), indicative of the presence of H₂O₂ in mitochondrial outer and inner membranes (B), autophagosome-like structure membrane (C), and cell wall (D). Ch, Chloroplast; CW, cell wall; L, lipid drop; M, mitochondrion; N, nucleus; P, peroxisome. (E) ROS production induced by C2-ceramide treatment of protoplasts. Protoplasts were isolated from 18-d-old wild-type leaves, treated with 0 (Control) or 30 μM C2-ceramide (C2) for 1 h under light, and then double stained with CM-H₂DCFDA (to detect ROS) and CMXRos (mitochondrial marker). Chloroplast autofluorescence was excited at 488 nm and detected at 738 to 793 nm (shown in blue). Images were taken by confocal laser scanning microscopy. At least 300 protoplasts were observed, and over 80% of protoplasts showed ROS signal. See Supplemental Figure 2 for the effects of different treatments on ROS production. This experiment was repeated three times with similar results. (F) and (G) Percentage of survival of wild-type protoplasts after ceramide treatment and coincubation of a ROS scavenger or inhibitors of protein kinase or protein synthesis, respectively. Protoplasts were isolated from 19-d-old wild-type leaves and then treated with chemical reagents under light. Cell viability was determined by fluorescein diacetate staining. Letters indicate statistically different values using a Student-Newman-Keuls t test, P < 0.05. For each treatment, at least 800 protoplasts were counted. Values are means ± sd from two independent experiments. (F) Effects of 5 μg/mL NAC on 30 μM C2 ceramide-induced cell death at 24 h. (G) Effects of 100 nM K252a and 10 μM cycloheximide (CHX) on 30 μM C2 ceramide-induced cell death at 5 h.

Figure 3.

Induction of Autophagy in acd5 Plants. (A) Expression level of autophagy-associated genes. The genes analyzed were ATG3, ATG7, ATG8a, ATG8e, ATG8g, and ATG8 h. Total RNA was extracted at the indicated days for quantitative RT-PCR analysis of gene expression. ACT2 encoding actin was used as an internal control. Gene expression values are relative to average wild-type levels of 22-d-old plants (set as 1). This experiment was repeated three times using independent samples. Values are means ± sd from three biological replicates. Significant differences were determined by Student's t test (*P < 0.05 and **P < 0.01). The primers used for these analyses are provided in Supplemental Table 1. (B) and (C) Detection autophagic vesicles inside the vacuole using GFP-ATG8e fusion. Different developmental stages of wild-type and acd5 plant leaves expressing a GFP-ATG8e fusion were infiltrated with 1 μM concanamycin A for 1 d. The detached leaves were visualized by confocal microscopy. Note abundant autophagosomes (arrowhead) in 40-d-old acd5 leaves. Bars = 10 μm. (D) Statistical analysis of number of GFP vesicles as shown in (B) and (C) in different developmental stages of wild-type and acd5 leaves. Twelve detached wild-type or acd5 leaves were observed in each developmental stage. Values are means ± se. Significant differences were determined by Student's t test (*P < 0.05 and **P < 0.01) from three biological repeats.

Figure 4.

Sphingolipid Content Alteration during acd5 Development. Sphingolipids were extracted from rosettes of soil-growth Arabidopsis leaves at different developmental stages. The amounts of free LCBs, main glucosylceramides, main hydroxyceramides, and main ceramides in wild-type and acd5 leaves were quantified after extraction, separation, and identification by HPLC coupled with electrospray ionization tandem mass spectrometry as described in Methods. Values indicate the absolute level of sphingolipids in acd5 relative to the level found in the wild type (set as 1) ±sd from four independent experiments. Significant differences between acd5 and the wild type at each time point were determined by Student's t test (*P < 0.05 and **P < 0.01) (see Supplemental Tables 2 and 3 for the absolute values). (A) Measurement of sphingolipids from acd5 relative to wild-type leaves at different times included total LCBs, glucosylceramides (gCer), hydroxyceramides (hCer), and ceramides (Cer). (B) Measurement of hydroxyceramide species in acd5 relative to wild-type leaves at different times. (C) Measurement of total long acyl chain length (C16, C18) and very long acyl chain length (>C18) ceramides or hydroxyceramides at different times in acd5 relative to the wild type. (D) Measurement of various ceramide species in acd5 relative to wild-type leaves at different times.

Figure 5.

Expression of Sphingolipid Metabolism Related Genes and Sphingolipid Content Alteration in acd5 and acd5 loh3 Double Mutants. (A) Expression of sphingolipid metabolism related genes in acd5 and wild-type plants. The genes chosen were three ceramide synthetase genes LOH1 (At3G25540), LOH2 (At3G19260), and LOH3 (At1G13580) and three candidate ceramidase genes: At1g07380, At5g58980, and At2g38010. Total RNA was extracted at the indicated time points for quantitative RT-PCR. ACT2 was used as an internal control. Gene expression values are presented relative to average wild-type levels of 22-d-old plants (set as 1). This experiment was repeated three times using independent samples. Values are means ± sd from triplicate biological repeats. Significant differences were determined by Student's t test (**P < 0.01). The primers used for these analyses are provided in Supplemental Table 1. (B) The acd5 loh3 double mutant phenotype. Photos are of 25-d-old plants. The white circles and the square indicate cell death lesions in acd5 loh3. Bar = 0.5 cm. (C) The transcript levels of ceramide synthetase genes in acd5 loh3 mutants. Total RNA was extracted from 22-d-old plants. ACT2 was used as an internal control. Gene expression values are presented relative to average wild-type levels (set as 1). This experiment was repeated two times using independent samples. Values are means ± sd from two biological repeats. Significant differences were determined by Student's t test (**P < 0.01). (D) and (E) Sphingolipid levels in acd5 loh3 mutants. The accumulation values indicate the absolute level of sphingolipids in 25-d-old acd5, loh3, and acd5 loh3 relative to that found in the wild type (set as 1). (D) The amount of total ceramides (Cer), hydroxyceramides (hCer), and glucosylceramides (gCer). (E) Various ceramide species. Values represent means ± sd from two independent experiments (see Supplemental Tables 4 and 5 for the absolute values).

Figure 6.

ROS Production in Wild-Type and acd5 Leaves after B. cinerea Infection. (A) Percentage of germination of B. cinerea at the indicated times after spraying conidia on wild-type and acd5 leaves. Conidial germination was visualized using trypan blue staining and counted by microscopy. This experiment was repeated more than three times with similar results, and more than 18 leaves in the wild type or acd5 were counted at each time point. Letters indicate significantly different values using Fisher’s protected least significant difference, a post-hoc multiple t test (P < 0.01). Error bars indicate sd from three independent tests. (B) Lengths of hyphae on wild-type and acd5 leaves after B. cinerea infection. The lengths of hyphae were measured with Leica LAS software. A total of ∼30 to 50 hyphae per leaf and a total of six leaves in each genotype were measured in each experiment. Asterisks indicate a significant difference from the wild type using Student's t test (*P < 0.01). This experiment was repeated three times with similar results and more than 150 hypha were measured each time. Error bars indicate sd from three independent tests. (C) Accumulation of ROS in wild-type and acd5 leaves at indicated times after 2% glucose (control) or B. cinerea inoculation. Leaves were stained with DAB as described in Methods. The brown precipitate indicated DAB polymerization at the site of ROS production. The right panel shows quantification of DAB staining as the percentage and intensity of leaf area stained from 18 leaves per genotype measured using ImageJ software as described in Methods. Error bars indicate sd from three independent experiments. Asterisks indicate a significant difference from the wild type using Student's t test (P < 0.05). Bars = 1 mm. (D) to (M) H₂O₂ localization observed by electron microscopy using cerium chloride staining after inoculation with 2% glucose or B. cinerea. The left panels are wild-type leaves inoculated with B. cinerea for 24 h (D), 36 h ([G] and [J]), and 48 h (K). Note that heavy cerium deposits (white arrows) were observed in the cell wall at 24 to 36 h ([D] and [G]), but not in mitochondria. The middle and right panels are acd5 leaves inoculated with B. cinerea for 24 h ([E] and [F]), 36 h (H), and 48 h (I). Note the presence of cerium deposits in both the cell wall ([E], white arrow) and the outer membrane of mitochondria ([F], black arrows) in 24 h acd5 samples. Mock-treated wild-type (L) and acd5 (M) leaves at 72 h stained with cerium chloride. Note the lack of H₂O₂ from the beginning of control treatment up to 72 h ([L] and [M]). Arrows indicate depositions of cerium. Ch, chloroplast; CW, cell wall; DC, dead cell; EH, extracellular hyphae; M, mitochondrion; N, nucleus; V, vacuole. Bars = 2 μm in (E) and (H), 1 μm in (D), (J), (L), and (M), and 0.5 μm in (F), (G), (I), and (K). [See online article for color version of this figure.]

Figure 7.

Compromised Defense Responses in acd5 Plants after B. cinerea Infection or Chitin Treatments. (A) Quantitation of callose deposits in the wild type and acd5 after inoculation with B. cinerea at the indicated times. This experiment was repeated three times with similar results. At least 18 leaves of the wild type or acd5 were observed in each experiment. Significant differences were determined by Student's t test (*P < 0.05). Error bars indicate sd from three independent tests. (B) Cross sections of wild-type and acd5 leaves after 2% glucose (control) or B. cinerea inoculation for 72 h. Arrowheads indicate pathogen hypha and boxes indicate dead cells. Bars = 50 μm. (C) Quantitation of callose deposits in the wild type and acd5 after treated with 100 ng/mL chitin at 12 and 24 h. This experiment was repeated three times with similar results. At least nine leaves of the wild type or acd5 were observed in each experiment. Letters indicate significantly different values using Fisher’s protected least significant difference, a post-hoc multiple t test (P < 0.05). Error bars indicate sd from three independent tests. (D) and (E) Electron microscopy images of cell wall appositions of wild-type and acd5 leaves after B. cinerea infection for 72 h. Three different leaf samples of the wild type and acd5 were observed and two independent experiments were done with similar results. Note the large cell wall apposition (CWA) in wild-type with cerium deposits (D) and tiny CWA in acd5 with free cerium (E). No large size CWA were found in any acd5 samples. Arrows indicate depositions of cerium. Ch, chloroplast; CW, cell wall; CWA, cell wall apposition; M, mitochondrion; V, vacuole. Bars = 1 μm. (F) Expression levels of defense-related genes in response to B. cinerea in wild-type and acd5 plants. Defense-related genes assayed were a chitinase family protein (At2g43580), PRX33, PRX34, PR1, PAD3, and CYP71A13. Total RNA was extracted at the indicated time points for quantitative RT-PCR analysis. ACT2 was used as an internal control. Gene expression values are presented relative to average wild-type levels at 0 hpi (set as 1). This experiment was repeated three times using independent samples. Values are means ± sd from three biological replicates. Significant differences were determined by Student's t test (*P < 0.05 and **P < 0.01).The primers used for these analyses are provided in Supplemental Table 1. [See online article for color version of this figure.]

Figure 8.

Ceramide Accumulation in Response to B. cinerea in Wild-Type and acd5 Plants. Three-week-old wild-type and acd5 plants were sprayed with10⁷ spores/mL of B. cinerea and sampled at different time points for sphingolipid quantification. (A) to (E) Relative level of sphingolipids in infected wild type versus 2% glucose controls (set to 1). (F) to (J) Relative level of sphingolipids in infected acd5 versus infected wild type (set to 1). Measurements were of four major sphingolipid classes ([A] and [F]), ceramide and hydroxyceramide containing various LCBs ([B] and [G]), ceramide and hydroxyceramide containing various acyl lengths ([C] and [H]), individual ceramide species ([D] and [I]), and individual hydroxyceramide species ([E] and [J]). Significant differences were determined by Student's t test (*P < 0.05 and **P < 0.01) (see Supplemental Tables 7 and 8 for the absolute values). Means ± sd are shown from three independent experiments.